摘要
为了提高静止变频电源输出的电压波形质量,增强控制系统的鲁棒性,提出了基于自适应粒子群优化算法(APSO)优化模糊神经PID控制策略。利用改进的自适应粒子群优化算法优化模糊神经网络的前件、后件参数和单神经元优化PID参数,实现了控制器参数的自动调整。在MATLAB/SIMULINK环境下,对该策略控制下的静止变频电源控制电路进行了仿真。结果表明,与普通的模糊神经网络PID控制对比,引入改进的粒子群优化算法可以实现参数的全局快速寻优。优化后的模糊神经PID控制器具有良好的控制性能和自适应能力,很好地满足了系统的鲁棒性、快速性的要求。
In order to improve the output-voltage performance and robustness of static inverter power supply, a method of fuzzy neural PID controller optimized by auto-adaptive particle swarm optimization algorithm （ APSO） is presented in this paper. By improved APSO algorithm which has global search characteristic, control strategy optimized fuzzy neural network parameters and the PID controller parameters are optimized by single neuron so as to realize the param-eters of the controller adjusted automatically. Based on MATLAB/SIMULINK circumstance, the system is simulated in the Static Inverter circuit. Compare with the fuzzy neural network PID control, simulation results on optimized fuzzy neural PID control show that the control system has good performance of adaptive capacity, which meets the high ro-bustness and the rapidity requirements of the system.
出处
《电测与仪表》
北大核心
2014年第5期41-45,共5页
Electrical Measurement & Instrumentation
作者简介
吴家梁（1990-），男，汉族，广东清远人，助理工程师，在读硕士，研究领域为军用电源设备新技术研究。Email：wujia2538@126．com
樊波（1965-），男，汉族，山西洪洞人，副教授，硕士生导师，研究领域为地空导弹电力系统新技术研究。
